Hi all,
I tried to run the "surgestandard" benchmark with the default
configuration. Here is my command line:
*./build/ALPHA_FS/m5.fast ./configs/example/fs.py
--benchmark=SurgeStandard*
**
* * I also monitor terminals of both the server and the client system. The
simulation seems running well. The following is a snapshot of the messages
printed on the client's terminal.
* waiting for server...
server ready
running surge client...SURGE: Scalable URL Reference Generator
Running 2 clients with 100 conns/client for 1 seconds
SURGEmaster: 90054 objects in name sequence
SURGEclient 0: running 100 connections
SURGEclient 0: Object count 0
SURGEclient 1: running 100 connections
SURGEclient 1: Object count 10000*
* .....*
* SURGEclient 1: Object count 90000*
* SURGEclient0: Hit end of name sequence*
* SURGEclient1: Hit end of name sequence*
I expect that the simulation should stop right after the two clients hit
the end of the name sequence (in the surge-client.rcS file, an "m5 exit"
command will be executed after running surge). However, the client system
get stuck after that: the simulation still continues even three days after
the above messages was shown (the terminal of the client system does not
print anything more, but hanging there). Could anyone have clue on the
reason of this? I didn't change anything of the simulator (source code,
configuration, rcS file...).
One more question is that where I can set the number of objects in name
sequence (90054 shown above). I know the number of clients, connections per
client and running time can be set in the surge-client.rcS file, but I don't
see the corresponding entry for the number of objects.
Thanks very much,
Ying
_______________________________________________
m5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
